This New England-based case-control study has sought to identify risk factors for ovarian cancer mediated through depletion of ovarian germ cells and gonadotropin excess. New theories with a molecular basis and considering roles for ovarian steroids, inflammation, and certain genetic polymorphisms now need to be incorporated. In our new model, gonadotropins remain important with risk factors operating at the pituitary or hypothalamic level. Elements within the ovary include germ cells regulating pituitary feedback, stromal cells producing steroids and proteins, and epithelial cells responding to growth factors. Epithelial cells trapped in inclusion cysts or deriving from endometriosis may be more susceptible to local or circulating growth factors. Epithelial proliferation in the setting of reactive oxygen species (ROS), generated from catechol estrogen or inflammation, leads to DNA damage. Selection of cells destined to become cancer may occur when inherited DNA repair mechanisms are faulty and oncogenes or tumor suppressors are damaged. Other genetic factors include variants of genes in pathways for steroid production or xenobiotic metabolism, while a variety of exposures may impact on risk. In the continuation proposed, existing and new epidemiologic data and DNA from 1700 controls and 1800 cases with ovarian, fallopian tube, or primary peritoneal cancer (including a subset identified pre-operatively) will be used to examine how hormonal factors, genes, and environment interact in this model. Through pituitary effects, pregnancy and lactation protect better than pregnancy alone and opposed regimens of menopausal therapy protect better than unopposed regimens. As a marker for poor germ cell reserve, short cycle lengths in very early reproductive life increase risk for ovarian cancer. By affecting steroid production, genetic variants of 3 and 17betaHSD, CYP11A, CYP17, and CYP 19 may affect risk for ovarian cancer. Steroid production is also enhanced by caffeine use, a predicted risk factor interacting with dietary cholesterol or caffeine metabolism. Genetic polymorphisms affecting the production or clearance of catechol estrogen or ROS generated by inflammation also influence ovarian cancer risk, including Ahr, CYP1A1, COMT, MnSOD, GST, and NAT and may interact with exposures like genital talc. Protection from ovarian cancer is associated with a variety of antioxidants including lycopene, carotene, ginkgo, and indoles from Brassica vegetables as well as anti-inflammatory agents. Our continued goal is to identify modifiable risk factors and chemo preventive agents for a common and lethal cancer in women.